IDEOLOGY

group would be any less complex. We will have to return to this ques­tion at the end of the third chapter.

4 IDEOLOGIEKRITIK

The members of the Frankfurt School hold three theses about Ideolo­giekritik:

1. Radical criticism of society and criticism of its dominant ideology (Ideologiekritik) are inseparable; the ultimate goal of all social re-­s.earch should be the elaboration of a critical theory of society of which Ideologiekritik would be an integral part.

2. Ideologiekritik is not just a form of 'moralizing criticism: i.e. an ide­ological form of consciousness is not criticised for being nasty, im­moral, unpleasant, etc. but for being false, for being a form of delu­sion. Ideologiekritik is itself a cognitive enterprise, a form of knowledge.

3. Ideologiekritik (and hence also the social theory of which it is a part) differs significantly in cognitive structure from natural science, and requires for its proper analysis basic changes in the epistemological views we have inherited from traditional empiricism (modelled as it is on the study of natural science).

In this section I will discuss various ways in ~hich Ideologiekritik might proceed, with particular attention to the questions: (a) In what sense is the particular kind of Ideologiekritik under discussion cognitive? (b) In what sense would a proper account of the kind of Ideologiekritik under discussion require revisions in our inherited epistemology? The forms of Ideologiekritik I will discuss in this section will all focus very nar­rowly on one of the three modes of criticism.

1. To begin with the first mode of Ideologiekritik - criticism along the epistemic dimension - to what extent can this kind of criticism be ac­commodated within a traditional empiricist framework? The members of the Frankfurt School take what they call 'positivism' to be the most consistent, plausible modern version of empiricism.

The Frankfurt School's 'positivist' begins by identifying:

(a) those statements or propositions which are potentially true or false; (b) those ,statements or propositions which have 'cognitive content' (i.e.

which, if true, would be 'knowledge'); (c) those statements or propositions which can be rationally assessed

(i.e. which are warrantedly acceptable or rejectable).

IDEOLOGIEKRITIK

Statements without 'cognitive content' are not true or false, but (cogni­tively) meaningless, and there is no sense in which they can be rationally discussed and evaluated. The positivist program gets its bite from its second step in which it identifies (a) through (c) above with:

(d) those statements or propositions which are. scientifically testable; (e) those statements or propositions which have observational

content.55

The identification of (d) with (a) through (c) may be called 'scientism' -roughly, the view that the only rationality is scientific rationality;56 the identification of (e) with (a) through (c) means that only statements with observational content are even potentially 'knowledge,' and that only they are subject to rational discussion and evaluation.

So the positivist, when confronted with a form of consciousness, can subject it to two kinds of criticism:

(a) scientific criticism: reject those beliefs in the form of consciousness which are empirically false or not well-supported;

(b) 'positivist Ideologiekritik': separate clearly 'cognitive' from 'non­cognitive' beliefs; reject all (second-order) beliefs which attribute to non-cognitive beliefs cognitive standing.

Of the four modes of epistemic criticism·7 (the positivist would claim) objectification mistakes and self-fulfilling beliefs will fall to (a) above: an objectification mistake is an empirically false belief - it falsely takes a state of affairs to hold simpliciter, which in fact holds only conditionally upon a particular kind of human social action - and a self-fulfilling belief is not well-supported - its evidence is tainted, but the taint can be discovered by further empirical investigation. The third of the four modes - confusion of epistemic standing - clearly falls under (b) above, but the fourth, confusion of a particular for a general interest, seems quite beyond the scope of positivist criticism.

None of the members of the Frankfurt School thinks that the tasks circumscribed by the positivist are insignificant ones - it is important that people not accept beliefs which are factually erroneous and don't take normative statements to be descriptive - but the positivist's notion of 'rationality' is too narrow and restricted, and can't handle any of the more interesting cases of ideological delusion; by excluding normative

"When the members of the Frankfurt School speak of 'the positivists' they have in mind primarily the Vienna Circle. But obviously the discussion in the text doesn't represent their views fully. The members of the Frankfurt School have no views on or interest in logic or mathematics, so I am going to ignore them in formulating 'positivism:

"Cf. El 13 [Tl 4f) . • 1 Vide supra, pp. 13f f.

IDEOLOGY

and metaphysical beliefs, preferences, attitudes, etc. from the realm of rational discussion and evaluation, the positivist leaves us without guid­ance about important parts of our form of consciousness, and thereby abandons whole areas of our life to mere contingent taste, arbitrary decision, and sheer irrationality.58 But how do we know that taste, pref­erence, and decision aren't the best we can do as guides to what atti­tudes, normative beliefs, etc. we should adopt? How do we know it isn't just wishful thinking to think that we can have some kind of normative knowledge, or attain some rational set of preferences and attitudes?

We are perhaps not yet in a position to say positively what the 'wider notion of rationality' which will have application to normative beliefs, attitudes, and preferences is, but Habermas believes that inspection of the actual practice of positivist 'Ideologiekritik' will show that positivists are tacitly using a more extensive notion of 'rationality' than they can admit. 59 Positivist Ideologiekritik operates, it will be recalled, by expos­ing non-cognitive beliefs (e.g. value judgments) which are masquerad­ing as cognitive. In itself this process need not lead to any changes in the substantive views of the agents. If I think that Haydn is a 'better' composer than Friedrich the Great and misconstrue this belief as a cog­nitive one, the positivist can enlighten me. I will then presumably learn that my original belief is actually a composite of: (a) a belief that ,the works of Haydn usually instantiate certain properties to a greater de­gree than do those of Friedrich the Great, and (b) a value judgment that those properties are the properties which make a work of music

.8TP 316-U [T 4 263-8]. This is the weakest (and also the most plausible) of three views critical of positivism to be found in the works of the members of the Frankfurt School:

(a) strongest view: positivism (and empiricism) do not give an acceptable account even of natural science;

(b) strong view: positivism gives a correct accOunt of natural science, but (given the nature of the subject-matter) is inadequate as an account of a theory of human society aimed at correct explanation and accurate prediction;

(c) weak view: positivism gives a correct account of natural science, and of the 'empirical-analytic' part of social theory (i.e. that aimed at correct explanation and predic­tion). but social theory a150 has a critical part - one aimed at something other than correct explaEation and prediction - and of this positivism can give 11<> account.

Marcuse holds (a); Habermas explicitly rejects (a) (TIN s0-60 IT5 82-91].) Natural science he simply gives away to the positivist. However he confuses (b) and (c). It isn't clear whether he is claiming:

(1) even to the extent to which natural and social sciences snaTt the common goal of explanation and prediction, the peculiar nature of the subject-matter of the social sciences means that these goals can be successfully attained only by the use of methods very differ­ent from those used in natural science;

(2) natural science and social theory differ radically in their cognitive goals. and so in their characteristic methods; or whether he is claiming some combination of both. In the text I hold to (c) .

• oTP 321 [T4268].

IDEOLOGIEKRITIK

'good.' Even if I were to accept the claim that (b) is a mere preference. utterly without any rational standing, I would still continue to prefer the. music of Haydn to that of Friedrich the Great.

But in some cases coming to know that I have made a mistake in the epistemic standing of a belief will cause me to give the belief up. In fact, in some cases there may be noncognitive beliefs, preferences, etc. which require to be accepted that they be mistaken for cognitive beliefs. But in what sense can a preference or a normative belief 'require' that the agents who accept it mistake it for a cognitive belief? What might be intended here are cases of the guidance or justification of action; we might think that for certain kinds of action only cognitive beliefs, not beliefs expressive of mere preferences, are acceptable as sources of guidance and legitimation. But if the only reason we hold the belief is that we (falsely) think that it is a cognitive belief. then when we are enlightened about its epistemic standing, we will give it up. After all, I may not share the preference expressed in the belief. Of course this is precisely the reason why anyone would bother to present a preference or value judgment disguised as a cognitive belief - agents will feel com­pelled to accept a true cognitive belief in a way in which they will not feel compelled to accept my preferences.6o If upon enlightenment I do give the belief up (because I don't share the preference on which it -is based), Habermas wants to say tha,t I have not just changed my be­liefs, but that I have moved to a more rational set of beliefs.

The effectiveness and significance of positivist Ideologiekritik de­pends on the fact that people do change their beliefs in the way de­scribed above; the positivists can cllUnt on people giving up beliefs which have been shown not to be cognitive, but to be expressions of prefer-

;.'

ences which cannot be acknowledged publicly as grounds for action. Positivist Ideologiekritik can have the right effect, then, but cannot give.­an account of its own activity in bringing that effect about. The moti­vation of the program must be to free agents from irrational belief and action by causing them to give up beliefs based on preferences those agents could not acknowledge; but the positivists can't admit that the motivation of the program is rational (since there aren't any 'rational motivations') or that the effect is to make the agents more rational. So

I. Note that this might be something like an 'objectification mistake' too. Our preferences. attitudes. etc. are somehow more the result of our own activity - we have more control over them than we do over what beliefs will be 'objectively true' of the world. Earlier members of the Frankfurt School were particularly terrified by fascism because they per­ceived it as opmly acknowledging that its politics was one of the nalc.ed exercise of power. based on arbitrary acts of will. This mode of ldeologiekritik. then. would have nothing to unmask.

IDEOLOGY

positivists can't justify their own activity of criticising ideologies except as a personal preference or an arbitrary decision.s1

To this the positivist may reply that the fact that people do change their beliefs as described in the last paragraph is no grounds for saying that they have thereby. become more rational, acquired a more justi­fied' or 't~uer" or more 'warranted' set of beliefs. What they have done is to bring their beliefs, preferences, and value judgments into closer agreement with the rest of their non-cognitive beliefs, e.g. beliefs about which preferences they 'ought' to allow themselves to express or by which they 'ought' to allow themselves to be moved. From the fact that the resulting set of beliefs, preferences, etc. is more coherent and con­sistent, it doesn't follow that it is 'knowledge,' or 'true.' Furthermore, it is sheer defamation to claim that positivists need consider their own activity a mere 'arbitrary' decision; to say that an activity is not grounded on some 'substantial concept of human rationality' (whatever that might mean) is not to say that it is based on some arbitrary decision. It isn't 'arbitrary' if it is motivated by deep-seated human needs, an expression of concern for human suffering, etc. But that doesn't make this decision one 'motivated by reason itself' - it is motivated by per­fectly understandable and unexceptionable human desires. The deci­sion to eat when one is very hungry is not arbitrary - I couldn't equally well have decided to go swimming - but that doesn't make eating a form of knowledge.

The task then for the members of the Frankfurt School is to give an account of what it means to say that the agents 'could not acknowledge' certain motives which shows how this means something more than that they in general don't like to acknowledge these motives, and to give an account of what it means to say that a belief 'requires' mistaken belief about its epistemic standing, which means more than just 'if the agents are enlightened about their mistake, they will injact give the belief up.' In what sense is it irrational to act on motives which one 'could not' acknowledge or to hold value judgments or preferences which 'require' mistakes about their epistemic standing?

In a way the oddest thing about this whole discussion is the extent to which Habermas is himself infected with the positivism against which he is struggling. In my reconstruction of the positivists' position I claimed that positivists tacitly identified statements which are poten­tially true or false (a), statements which have cognitive content (b), and statements on which there could be a rational consensus (C).62 The pos-

11 Cf.TP 11 20£ [T 4 267f] . •• Vide supra pp. 26f., infra pp. 88f.

IDEOLOGIEKRITIK

itivists then went on to argue that all rationality was scientific rationality (d), and that all scientifically meaningful statements were statements with observational content (e). Since attitudes, preferences, v,aluejudg­ments, normative beliefs, etc. obviously have no direct observational content, they can't be true or false, hence-they are cognitively meaning­less, so there are strong limits to rational discussion of them, and ulti­mately one can have no warrant for adopting or acting on them; any consistent set of preferences, attitudes, etc. is as good, as 'rational; as any other.

Habermas counters in the obvious way: Clearly not any consistent set of preferences, attitudes, and normative beliefs is as 'rational' as any other. This sense of 'rational' may be unclear and difficult to analyse, but that doesn't mean that it is illicit or doesn't exist, and if positivism can't give an account of it, so much the worse for positivism. But instead of going on to attack the first part of the positivists' view - the identifi­cation of (a) through (c) - Habermas accepts it; so, if some normative beliefs are more rational than others, there must be a kind of normative knowledge; because some preferences and attitudes are more rational than others, sets of preferences and attitudes can be 'true' or 'false.' 'Truth' and 'falsity' as used in science do not admit of degrees; a prop­osition is true or false, and tertium non dntur. But rationality is not like that. Decisions, preferences, attitudes, etc. can be more or less rational; agents can have stronger or weaker warrant for their actions, can be more or less aware of their own motives, can be more or less enlight­ened in their normative beliefs. If I am asked whether I agree with Habermas or with the positivists, that is, whether I think that there is a single, 'true; uniquely rational set of human preferences, attitudes, and normative beliefs, or whether I think that there is no sense in which any set of attitudes, preferences, and normative beliefs is 'more rational' than any other, the only reasonable reply is to reject this alternative as falsely posed.

II. With that I turn to the second approach to Ideologiekritik, that in terms of the functional properties of forms of consciousness. An ideol­ogy is a world-picture which stabilizes or legitimizes domination.s3 But

.3 Vide TG ~45. ~47. ~57, ~59. 279. ~85f. ~89f. I won't discuss the second and third func­tional senses of ideology, i.e, 'an ideology is a form of consciousness which serves to mask social contradictions: and 'an ideology is a form of consciousness which hinders maximal development of the forces of production (cf. supra. p. 18.)

In some cases 'masking social contradictions' may be a way of supponing or legitimizing an oppressive social order. so a form of consciousness which is an ideology in that it masks social contradictions will also be an ideology in the sense that it suppom or legitimizes 'Herrschaft' .

32 IDEOLOGY

what is the relation between the 'falsity' of the form of consciousness and its functioning to support or legitimize oppression?~4 There are four possibilities:

A. The world-picture is false and it stabilizes or legitimizes oppres~ sion, but its falsity and its oppressive functioning have no inherent con­nection. We can know tha.t it is false independently of knowing w~ether it functions oppressively, and we can know that it functions oppressively independently of knowing whether or not it is false.B5

B. The world-picture is false - we assume from the start that we have whatever grounds are necessary for asserting that - and the judgment that the world-picture functions oppressively is parasitic on our judg­ment that it is false. Consider the following case: Suppose we have a world-picture, a central part of which is a set of normative beliefs which are used to give legitimacy to the basic institutions of the society, Like most social institutions, these will be likely to have repressive features; the question then is whether this is necessary or legitimate repression, or whether it is oppression, domination, Herrschaft, etc. One way to answer this question would be to look at the arguments given for the legitimacy of the institution; if these arguments are correct and start from 'true' normative beliefs, the institution (and the repression asso­ciated with it) is justified. if the 'best' available arguments must start from false normative beliefs, the institution is not justified, and, if in addition it imposes frustration of given human preferences, it is a form of oppression, So to know whether what the world-picture supports and legitimizes is 'Herrschaft' or oppression one must know whether the world-picture is itself true or false. .

This answer won't work, of course, but it points in the right direction. From the fact that the best arguments the members of the society can

•• As section ~, part 11 of this chapter indicates there are various quite different views about what exactly it is that an ideology in the functional sense stabilizes Or legitimates: repres­sion, Herrschaft, illegitimate repression, ZwangsverhiHtnisse (TG ~4 7), surplus repres­sion, etc. I will use the term 'oppression' and its derivatives (e.g. 'functions oppressively') in the following as a place-holder for whatever reprehensible feature of society 'ideology' is defined as stabilizing and legitimating .

•• At TP 310 [T 4 2571 Habermas speaks of ideological error in a way which combines all three approaches to ideology: 'Der lrrtum, mit dem es die Aufklarung zu tun hatte, ist vielmehr das falsche Bewulltsein einer Epoche, das in den I nstitutionen einer falschen GeselIschaft verankert ist, und ihrerseits herrschende Interessen befestigt.' So ideological consciousness is:

(a) epistemically false (i.e. it is an 'lrrtum'); (b) functionally reprehensible (i.e. it 'befestigt herrschende lnteressen'); (c) genetically unacceptable (i.e. it is 'in den lnstitutionen einer falschen Gesellschaft

verankert'). Although it is hard to be sure, it would seem as if Habermas means here that these three things are independent of each other.

IDEOLOGIEKRITIK 33

give are not valid, it doesn't follow that there are no valid arguments to the legitimacy of the institution. So the judgment that this world-picture is oppressive is not parasitic on a judgment that it is false, but it may depend on a claim that no true world-picture could yield valid argu­ments for the legitimacy of the institutions, or - if one wishes to adhere strictly to the principle that Ideologiekritik is a form of 'internal criti­cism' - that no true world-picture 'acceptable' or 'accessible' to the agents could yield valid arguments for the legitimacy of the social insti­tutions.66

C. We can reverse B and give philosophical primacy not to the epis­temic properties of the world-picture, but to its functional properties; A world-picture or form of consciousness is 'false' in virtue of the fact that what it stabilizes or legitimizes are relations of Herrschaft.

It is hard to see how this could work in this simple form. A form of consciousness might contain (among other things) some simple descrip­tive beliefs; if these are observationally accurate, why.call them 'false,' whatever their functional properties. To use 'false' in this 'functional' sense of descriptive beliefs could cause nothing but confusion; the same belief might turn out both (observationally accurate and hence) 'true' and (functionally) 'false.' But'I assume that the whole point of the use of words like 'true' or 'false' is that they represent a definitive judgment on whether a belief is to be rejected or accepted,67 so we want to avoid cases in which we say of the same belief that it is both true and false.

Still we may be able to salvage this 'functional' sense for a world-pic­ture or form of consciousness as a whole. If one thinks of the charac­teristic components of a form of consciousness as attitudes, preferences, and normative and metaphysical beliefs. one might claim that these have no clear observational content to get in the way. We might wish to express a definitive judgment on their acceptability by calling them

.. With this terminological distinction between 'acceptable' and 'accessible' I want to signal a difference between what the agents would adopt if it were presented to them, and what they could reasonably be expected to develop themselves in their historical situation.

"Vide supra pp. 31ff, infra pp. 94ff. Some members of the Frankfurt School, notably Adomo (vide my review of Negative DiahlUik in theJouf1Ul1 of Philosophy, 1975), may'hold the Hegelian view that a statement or form of consciousness can be both true and false. Fortunately I need not try to explicate this difficult and obscure doctrine, because, as his paper on truth (WT) makes clear, Habermas does not wish to adopt this usage. Even in his early works in whiCh the traces of Hegel are most marked, Habermas avoids saying directly that ideological consciousness is both true and false. Thus in the book re~ie,:" 'Li­ieraturbeticht zur philosophischenDiskussion urn Marx und den Marxismus' published in 1957 (reprinted TP 387ff) he says that ideological consciousness may be a 'correct' or 'accurate' representation of reality, but is 'false:' 'Es selbst (scil. das Bewu8tseinJ wird falsch durch die - sogar richtige - Spiegelung einer falschen Wirklichkeit .... {die Ideologie findetl ihre au8ente Grenze in der korrekten Abbildung des falschen Bestehenden' (TP 437). Cf. O'Neill, p. -S6.

34 IDEOLOGY

'false' if they stabilize or legitimize relations of 'Herrschaft'; otherwise, true. We needn't worry that this will conflict with our judgment on their descriptive accuracy; there won't be any. So the judgment that a world­picture or form of consciousness is 'ideologically false' (on this view) is ajudgment passed on the form of consciousness as a whole in virtue of the way in which the characteristic attitudes, preferences and normative and metaphysical beliefs which make up the form of consciousness function in the society.68

One might take it as a great advantage of this approach that which forms of consciousness are ideologically false wi\l change over time and depend on details of the particular historical situation; the same set of beliefs and attitudes may in some historical epochs support repressive social structures and in others may not.69 Thus, what were originally 'idle' metaphysical speculations about the nature of'o'v<i~a' with no social function at all, save perhaps to provide Greek gentlemen with a post­prandial conversational alternative to the flute-girls,70 may in a differ­ent historical context be absorbed into Christian theology and acquire a repressive function.

If, however, we are going to use the reprehensible functioning of a world-picture to define its ideological falsity, we must have a very dear idea of which modes of functioning are reprehensible. This brings us back to the old question: Is the form of consciousness reprehensible (i.e. ideological) if it functions to support or legitimize Herrschaft, sur­plus repression, surplus Herrschaft, illegitimate repression, etc.?71

'Herrschaft' the reader will recall, was defined relative to the ability to frustrate agents' wants and preferences. 72 'Macht' (power) is the abil­ity to impose on agents the frustration of their preferences, or, as Ha­berm as puts it in Theorie der Gesellschrift, Macht is the ability to prevent others from pursuing their interests.73 The exercise of Macht is repres­sion. This exercise is either 'manifest' - open use of force or direct threat to use force - or 'normative.' Repression is 'normative' if the agents are prevented from pursuing their interests by' a set of norma-

•• .. One needn't even deny that the form of consdousness might contain some simple de­scriptive beliefs; these remain true or false in the normal empirical sense regardless of the 'ideological' judgment on the form of consciousness as a whole.

"The same is true of other 'functional' approaches, i.e. of the definition of ideology as what serves to hide social contradictions and the definition of ideology as what hinders maximal development of the forces of production.

'"The avll.,x of course was not a flute - it had a double reed - but in this case mistranslation may have aesthetic advantages. Imagine Alkibiades' entrance at the end of the Symposium 'drunk, crowned, and supported by a female oboist.'

"Vide supra pp. 15-18. "Vide supra p. 16. 13TG 254.

IDEOLOGIEKRITIK 35

tive beliefs they accept. 'Normative power' if distributed 'asymmetri-cally' is 'Herrschaft.' "

The appeal of the notion of 'surplus repression' (and of the notion of 'surplus Herrschaft') derives from the hope that it would provide us with an 'objective' standard for evaluating societies and their associated forms of consciousn'ess - we might even be able to measure the 'surplus repression' in a given society at least approximately and comparatively. But to measure the amount of 'surplus repression' in a society we would have to determine what the wants, needs, preferences, and desires of the members of the society are and what the economy 'requires' to be able to reproduce itself. Unfortunately both of these notions - the 'wants, needs, and desires of the agents' and the 'requirements of the economy' - are highly problematic.

To say that ' ... is required by the economy; or ' ... is required if the economy is to function' is at best an ellipsis; to say that 'the economy' 'functions' is to say that it functions at some level of efficiency (i.e. at some level of expenditure of resources) to provide some level of satis­faction of a qualitatively specific set of human wants and needs of a quite particular kind; it presumably also means that the industrial plant does not disengorge a mass of goodies and then promptly collapse, so the economy must function in such a way that it not only satisfies certain human wants, but also 'reproduce itself.' One can't even begin to determine whether the society imposes surplus repression on its mem­bers unless one specifies what level of satisfaction of what particular hu­man desires and needs the economy is to provide. Did the ancient econ­omy 'require' the use of slave labor in order to function at a level which would provide each Mikrokles with an onion, a barley cake, and half a cabbage for each of his two me"als a day? Or did it require the use of slave labor if it was to function so as to allow Alkibiades to race in a four-horse chariot in the morning and spend the afternoons playing cottabos in the baths? Or would Mikrokles get only one onion a day if Alkibiades spent less time racing? The extra onion then is Mikrokles' share of the 'fruits of Athenian imperialism.'

Associated with every human society there will be a set of 'accepted' wants, 'needs: and desires, and a traditional level of expected satisfac­tion of these wants and desires. 74 But one can't define 'surplus repres­sion' as any repression more than that required for the economy to function so as to satisfy the socially accepted wants and 'needs' of the agents at the traditional level. After all, the set of 'accepted' wants, needs, and desires, and the traditional level of consumption may them-

'·Vide Sahlins (1976), Chapters 2 and 3.

IDEOLOGY

selves well be part of the 'ideology' we wish to criticize; appeal to 'sur­plus repression' was supposed to give us a standpoint IlUtside the given social interpretation of the agents' needs, from which to criticize it. On the other hand, no member of the Frankfurt School wishes to argue that the only 'real' wants and needs are those which must be satisfied to insure minimal biological survival, so that any repression more than that needed to insure the physical survival of the agents is 'surplus.' Agents in society have acquired sophisticated cultural needs, and it is as legitimate and important that they be satisfied as it is that the agents get enough food and shelter. a

This suggests that 'surplus repression' can't be the starting-point and basis for Ideologiekritik. Rather it seems that one must start with some kind of theory of which needs and wants are legitimate and which ide­ologically false; the amount of surplus repression, then, is the amount of repression exacted beyond that required to satisfy the agents' legiti­mate wants and needs. 'This repression is surplus' is a conclusion which sums up a critical argument; the real work of ldeologiekritik will al­ready have been done in distinguishing legitimate from 'false' wants and needs.

Perhaps we can still find a way to give philosophical primacy to the functional properties of a world-picture, but the notion of 'surplus repression' won't give us a quick and short way to an 'objective' defini­tion of 'oppression.'

D. Finally we might claim that neither the falsity of the world-picture, nor its mode of functioning, have any primacy in the analysis of ideol­ogy because they are inherently interconnected. One can't determine that the world-picture is false apart from an argument which at some point crucially appeals to the fact that this world-picture supports or legitimizes admitted oppression, nor can one show that what the world­picture supports or legitimizes is in fact oppression without some appeal to the falsity of the world-picture. Since Peirce and Dewey such contex­tualist views are not completely unknown among English-speaking phi­losophers; members of the Frankfurt School would probably call this approach 'dialectical.'

Ill. The third approach to Ideologiekritik is in terms of the genetic properties of forms of consciousness. How can a form of consciousness be 'false' in virtue of something about its origin, history, or genesis?

One widely practiced form of genetic Ideologiekritik is what I wiII call the 'social origins approach.' In certain forms of vulgar Marxism to

IOEOLOGIEKRITIK 37

call a belief a 'bourgeois belief' or a 'feudal belief' is to criticize it as ideologically false because of its social origin, i.e. because it typically or characteristically arises in societies dominated by the bourgeoisie, or because the original proponents of the belief are themselves members of the bourgeoisie.

Prima facie this doesn't seem to be a very promising line of argument. Most nineteenth- and eariy-twentieth-century natural scientists were working in a society dominated by the bourgeoisie, and perhaps one might argue that their physics was a kind of intellectual enterprise that characteristically arises in bourgeois societies, obsessed as they are with efficient and reliable methods for controlling natural phenomena. Still, although we may explain why particular physicists came to hold a par­ticular false physical theory by reference to their social environment and the 'requirements of bourgeois society,' the theory is not false be­cause it arose in a bourgeois society, but because it is inaccurate, incom­patible with the evidence, etc. Why should the situation be any different for 'forms of consciousness?'

Proponents of the social origins approach think that the situation is different for forms of consciousness because they hold a very strong view about the connection of the social origins of a form of conscious­ness and certain other facts about that form of consciousness. They believe that if a form of consciousnesness characteristically arises among the members of a certain social class, that form of consciousness will be the 'expression' of the (cIass-) position, standpoint, or 'view­point' of that class, or, alternatively, that, if a form of consciousness characteristically arises in a society dominated by a particular social class, it will be an 'expression' of the class-position, or standpoint, or viewpoint of that dominant class. 76 To say that a form of consciousness is an expression of the position in society or viewpoint of a social class is.to say:

(a) that the form of consciousness formulates the class-interests of that social class,

or,

(b) that the form of consciousness represents social reality as it appears to the members of that social class,

or,

(c) both (a) and (b).

IDEOLOGY

It is important in (a) above that when we speak of a form of conscious­ness as the 'expression' of class-position, we mean that it Jormulates' the class-interests of that social class, and not merely, for instance, that adopting and acting on that form of consciousness will further the class­interests of some social class. To say that acting on a particular form of consciousness will further the interests of some social class is to attribu{e to that form of consciousness a 'functional property: Perhaps one might want in fact to claim that, if a form of consciousness characteris­tically arises among the members of some class, acting on it will usually in fact foster their class interests, but this may turn out to be a form of 'functional criticism' in disguise, that is, one may not be criticizing the form of consciousness because it arises thus and so, but because given that it arises thus and so it will have these and those functional proper­ties.

What is wrong, though, with 'formulating' or 'fostering' a class-inter­est; why is this grounds for rejecting a form of consciousness as ideo­logical? It might seem grounds rather for cherishing it. Why isn't a form of consciousness that correct!y formulates the interests of a social class an ideology 'in the positive sense?'17

The answer is that, although the class-interest of some particular class may in some circumstances be identical with the 'general interest' of the society, very often this will not be the case. But classes have a natural tendency to identify their own particular class interest with the general interest. A form of consciousness is to be rejected if it falsely presents a particular class-interest as the general interest of the society, or if, al­though purporting to foster the general interest, it in fact fosters the particular interest of a social class. 78

If this, or something like it, is what is intended by those who criticize a form of consciousness for being 'an expression of the position in so­ciety of a particular class,' namely that this form of consciousness falsely presents the particular interest of this particular social class as the gen­eral interest, this is not a form of genetic Ideologiekritik. The form of consciousness in question is not being criticized in virtue of its origin, but in virtue of the 'falsity' it is likely to have as a result of this origin. Its origin in the particular experiences of a particular social class will then be at best a more or less reliable indicator that the form of con­sciousness will be found to present a particular class-interest as the gen­eral interest. So this purportedly genetic Ideologiekritik is actually a kind of epistemic Ideologiekritik.

Similar conclusions follow if one interprets the statement: 'This form

77Vide supra pp. 22-6. 78 LS S8C, ISSff [Ta 22f, 1I1ff].

IDEOLOGIEKRITIK 39

of consciousness is an expression of the viewpoint of this particular social class' as meaning This form of consciousness represents social reality as it appears to the members of that social class' (as in (b) above). That a form of consciousness represents social reality as it appears to the members of some social class would seem no grounds for rejecting it. But suppose one holds that the 'truth' about society is a 'total' view of that society, i.e. some kind of combination or integration of the ways the society appears from the perspective of each of its constituent groups. Then a form of consciousness which represents social reality as it appears to the members of some particular social class is a merely partial view of the society, i.e. it is not the 'truth' about the society, i.e. it is false consciousness.79 Fortunately it is not necessary to try to cash in the metaphors of this dubious theory of social knowledge; here too, clearly, what is at issue is a form not of genetic but of epistemic Ideo­logiekritik. What is wrong with ideological forms of consciousness is not their origin, but their false representation of social reality.

So far we have not found any genuinely 'genetic' form of Ideologie­kritik. Perhaps we will fare better if we try to take seriously the analogy between Ideologiekritik and psychoanalysis, and between ideologically false consciousness and individual neurosis.80

In Die Zukunft einer Illusion Freud distinguishes between 'error' ('Irrtum'), 'delusion' (,Wahnidee'), and 'illusion' ('lIlusion').81 An 'Irrtum' is just a normal, everyday, false factual belief, e.g. the belief that Sigmund Freud was born in Vienna is an 'Irrtum.' A 'Wahnidee' is a false belief an agent holds because holding this belief satisfies some wish the agent has: e.g. a man who falsely believes that he is Charle­magne because this belief satisfies his wish to be an important historical personage is suffering from a delusion, a ·Wahnidee.' An 'illusion' is a belief which mayor may not be false, but which is held by the agent because it satisfies a wish. Freud's example of an 'illusion' is the belief of a middle-class girl that a prince will come and marry her. It may in fact turn out that a prince does come and marry her - in Freud's Vi­enna there were such princes around, although probably not very many, so the girl's chances were rather slim - but the reason she be­lieves that she will marry a prince is that this belief satisfies some wish she has.

The 'illusion' mentioned in the title of Freud's work is religious belief, but his discussion is not as clear and unambiguous as one might wish. Some religious beliefs are like 'Wahnideen' - patently false beliefs to

7. Mannheim, 28~ff, I03ff . • OTW 15gf[Tl 311f). 81 Freud, pp. 164f.

IDEOLOGY

which the agents cling because they satisfy some deep needs - but most religious beliefs are merely 'illusions' - beliefs of indeterminate truth­value which are accepted because they satisfy agents' wishes.

But are religious beliefs really 'iIIusions' in the way in which the mid­dle-class girl's belief about 'her' prince is an illusion? The middle-class girl's belief is of indeterminate truth-value - it mayor may not be true - only in the sense that she· does not now have. any evidence for it . .But in itself the belief is true or false - either the prince will come or he won't. Freud himself suggests that this is not quite the case with reli­gious beliefs: 'Cber den Realitiitswert der meisten von ihnen kann man gar nicht urteilen. So wie sie unbeweisbar sind, sind sie auch unwider­legbar.'82 This might be taken to mean: We can't prove them or disprove them; but we could in principle have evidence for them - they are in themselves true or false - it is just that at the moment we don't have any evidence for them either way, just as the girl has no evidence for her belief. But it might also mean: We can't make any judgment at all about them as representations of reality; they are so vague and unspecific, we wouldn't know what to count as evidence for them - perhaps it is even wrong to think of some of them as purporting to represent reality at all rather than merely expressing certain attitudes. What could be the evi­dence for a belief that 'there's a destiny that shapes our ends, rough­hew them as we will?' Or for the belief that 'All there is, is either sub­stance or attribute of substance and God is the one substance?' From here it is but a step to the claim that there is no point in calling such things 'true' or 'false: Preponderance of the evidence cQUid be the rea­son the middle-class girl thinks she will marry a prince (but it probably isn't). but no one could hold these religious beliefs as a result of consid­ering the non-existent or completely inadequate evidence for them. so the reason these beliefs have been able to perpetuate themselves through millennia must be that they satisfy agents' needs and wishes.

So we must distinguish (at least):

(a) cases of delusion: Despite overwhelming evidence that the belief is false, the agent continues to hold it because it satisfies some wish;

(b) cases of illusion in which the belief is one for which the agent c&uld have adequate evidence, but' which is accepted by the agent because it satisfies some wish;

(c) cases of illusion involving beliefs for which there could not be ade­quate evidence (and which therefore must be accepted because they satisfy some wish).

'Z Freud. p. 165.

IDEOLOGIEKRITIK

What are we to say, then, about ideology in the pejorative sense? Is it a kind of delusion or of illusion? Let me start with what I will call the generalized 'wishful thinking model' of Ideologiekritik. According to this model, Ideologiekritik proceeds by showing:

(a) that certain agents make a characteristic kind of mistake Cb) that one can explain why they make t,hat mistake by ultimate refer~

ence to interests.

This model differs from normal cases of 'wishful thinking' and from the cases Freud discusses in that the explanation in (b) need not be an explanation in terms of individual psychology. Thus the agents may make the mistake because of the institutional context within which they act, and we may explain why this institutional context has the charac­teristics it must have in order to produce the mistake by reference to the interests of some agents, but those who make the mistake may not be motivated as individuals to satisfy the interest by reference to which the mistake is explained. In fact, it may not even be the case that the interest by reference to which the mistake is explained is an interest of those who make the mistake. Consider, for instance, members of some disadvantaged group who are employed in a government bureau to collect and analyse unemployment statistics. Let us suppose that the rate of unemployment in the society is systematically underestimated and that the ultimate explanation of this fact is that this underestima­tion is in the interest of some powerful group in the society. The agents in the government bureau who 'make' the mistake may have no interest whatever which is satisfied by holding a false belief about the rate of unemployment - in fact underestimation of the rate of unemployment may be directly contrary to their interests. The way the 'interest' trans­forms itself into error is not by providing them with a direct individual incentive or motive to make the mistake, but by arranging the condi­tions under which the statistics are collected and evaluated so that ra­tional agents working in those conditions will in general be prone to make this kind of systematic error.

The 'generalized wishful thinking niodel' described above makes ide­ology out to be a kind of 'delusion: a 'Wahnidee.' The belief in question is one which is clearly false, there is ample evidence available for ra­tional agents to see that the belief is false, but they don't, and the reason they don't is that powerful interests are operating to place them in non­standard conditions.

It is, of course, an extremely important task for empirical social re­search to point out how the interests of powerful social groups cause false information to be produced and disseminated throughout the 50-

42 IDEOLOGY

ciety, but it is not the task of Ideologiekritik. In the case described above - as in typical cases of what Freud called 'delusion' - the mistake or error in question was a straightforward factual error. But no amount of factual error is in itself sufficient to render a form of consciousness ideologically false. If a form of consciousness is ideological, one result might be that certain kinds of truths were systematically overlooked or certain kinds of errors systematically made. That certain kinds of errors are characteristically made, might lead us to suspect that some 'ideolog­ical' element in the form of consciousness would be threatened by cor­rect belief, but this does not mean that ideological falsity consists in fac­tual ignorance or false factual belief. On the 'genetic' view under consideration in this section, the ideological falsity of a form of con­sciousness is supposed to consist in something about its origin or gene­sis; Ideologiekritik was to ferret out this peculiar kind of non-empirical error. But neither part of the criticism along the lines of the model of wishful thinking - neither (a) nor (b)- seems, to require any but the normal empirical methods of social research, and what is wrong about an underestimation of the rate of employment is not that it has any particular origin, but that it is an underestimation, i.e. that it is factually in error.

So, it would seem, if ideological error is 'delusion' in Freud's sense, or is appropriately analysed in the 'model of wishful thinking' in the unspecific form in which that model has been presented above, Ideo­logiekritik is not an activity which requires any revisions in received views about epistemOlogy.

If ideological error is taken to be like the illusion from which the middle-class girl in Freud's example suffers, the same argument would seem to apply. One doesn't show that the belief is false by showing that this person holds it because it satisfies one of her wishes. The way to criticise the belief is not to show that she wishes it to be true, but to show how inherently implausible it is, and we do that by 'normal empirical means.' To this it might be objected that the point here is not to show that the belief is false, but to criticize the agent for adopting an inher­ently implausible belief for which she had no evidence. But it requires no major revisions in our epistemology to treat cases like this either; why is even the positivist estopped from criticizing agents for holding implausible, empirically unsupported beliefs?

Are ideologies, then, like the second kind of illusion, i.e. like (c) in my scheme above?83 That is, are ideologies forms of consciousness with little or no observational content, which, therefore, if adopted at all,

"Vide supra, p. 40.

IDEOLOGIEKRITIK 43

must be adopted because they satisfy some wish, desire or interest of the agents? It can't in itself be an objection to a belief that it satisfies a wish or desire, or even that it is accepted because it satisfies a wish or desire. True beliefs for which I have good evidence will satisfy my desire to accept true beliefs for which I have good evidence, and that they satisfy this wish is the reason I accept them. What is wrong with 'wishful think­ing' is not that we accept beliefs because they satisfy desires we have, but that we accept these beliefs because they satisfy the wrong, i.e. in­appropriate, desires. Empirical beliefs can be accepted because they sat­isfy our wish to accept well-confirmed empirical beliefs; if we accept them because they satisfy some other wish, we are engaging in 'wishful thinking.' This suggests that we might be able to distinguish appropri­ate and inappropriate motivations for different classes of beliefs. Even if the only appropriate and acceptable motivation for accepting empir­ical beliefs is the desire to accept only well-confirmed beliefs, this cannot be an appropriate motivation for accepting normative and metaphysical beliefs or for adopting attitudes, preferences, etc., since there is no way in principle in which any of these things CQUld be empirically confirmed.

It is just not an option for us as human beings not to have some atti­tudes, preferences, and normative beliefs. Is there some way, then, of distinguishing appropriate and acceptable from inappropriate and un­acceptable motives for attitudes, preferences, and non-empirical be­liefs? Should preferences, attitudes, etc. be rejected if they have been adopted for unacceptable motives? Are they then to be rejected as 'false?' Empirical beliefs adopted as the result of wishful thinking need not be false; in fact we generally speak of 'wishful thinking' only in ca,ses where the belief is false or at least very implausible. But if 'wishful thinking' just means (as we have taken it to mean) accepting a belief because it satisfies some wish (other than the wish to accept well-sup­ported beliefs), there is no reason why we might not have some evidence for a belief accepted because of 'wishful thinking; as long as the evi­dence is not the reason we accept it. So to criticize agents for indulging in wishful thinking is not necessarily to show that their beliefs are to be rejected as false. This line of argument can't be carried over to prefer­ences, attitudes, and non-empirical beliefs, because with them one can't distinguish their truth or falsity from the motives the agents have for accepting them. Still, this may mean no more than that non-empirical beliefs are not true or false at all (since they are not 'observationally' true or false); it certainly doesn't imply that we show non-empirical be­liefs, attitudes, preferences, etc. to befalse by impugning the motives of those who adopt them.

A prime example of the genetic approach to ideologiekritik is

44 IDEOLOGY

Nietzsche's criticism of Christianity.84 This criticism is 'genetic' because it appeals to a purported fact about the 'origin' of Christianity - that Christianity arises from hatred, envy, resentment, and feelings of weak­ness and inadequacy. To say that Christianity 'arises' out of hatred and envy is presumably not to make a historical statement - it is unclea:r what critical import such a statement could have - but to make a general statement about the typical motivation of Chris.tians. How do we know that these motives are 'unacceptable?' Nietzsche, in presenting this crit­icism, need not himself be committed to the view that hatred is in gen­eral, or always, or even ever an unacceptable motive for action. It is sufficient for the critical enterprise that the Christian cannot acknowl­edge hatred as an acceptable motive for beliefs, preferences, and atti­tudes. Since it is a central doctrine of Christianity that agents ought to be motivated by love, and not by hatred, resentment, envy, etc., Chris­tianity itself gives the standard of 'acceptability' for motives in the light of which it is criticized. If Nietzsche's account of its 'origins' is correct, Christianity 'requires' of its adherents that they not recognize their own motives for adhering to it. It isn't very important whether one wants to say that this criticism, if correct, shows Christianity to be 'false,' or 'merely' something else - contradictory, radically irrational, unstable, ete. The point is that the Christian who accepts Nietzsche's argument and thereupon gives Christianity up, is not doing anything analo&ous to acting on a whim, expressing a mere preference, or making an arbi­trary decision; this action is rationally grounded.

This example suggests that at least in some cases we can 'criticize' a form of consciousness because of the motives which lead the agents to adopt it. But the 'origin, genesis, and history' of a form of consciousness includes more than just the motives of the agents who adhere to it. Chapter 3 will treat a kind of genetic Ideologiekritik in which a form of consciousness is criticized because of non-motivational features of the conditions under which the agents could have acquired it.

s. Ni~tzsche (1969).